Environ Earth Sci (2017) 76:366
DOI 10.1007/s12665-017-6691-9
ORIGINAL ARTICLE
Evaluation of present-day rainfall simulations over West Africa
in CORDEX regional climate models
A. A. Akinsanola1,2 • K. O. Ogunjobi2
Received: 28 September 2016 / Accepted: 3 May 2017 / Published online: 13 May 2017
Ó Springer-Verlag Berlin Heidelberg 2017
Abstract The objective of this study is to evaluate the
ability of seven CORDEX regional climate models
(RCMs), driven by ERA-Interim reanalysis dataset to
simulate the observed rainfall characteristics over West
Africa during the period of 1990–2008. The seasonal cli-
matology, annual rainfall cycles, interannual variability,
850 hPa specific humidity, and wind fields of the RCMs
outputs were assessed over a number of spatial scales
covering three climatically homogenous subregions (Gui-
nea Coast, Savannah, and Sahel) and the entire West Africa
domain. The ability of the RCMs to simulate the response
to El Nino and La Nina events were further assessed.
Results indicate that the RCMs captured the spatial pattern
of rainfall and the three distinctive phases of the West
African monsoon reasonably. It is worth noting that RCA
and CRCM5 failed to distinctively reproduce the monsoon
jump while CCLM, HIRHAM, and REMO largely over-
estimated the amount of the pre- and the post-monsoon
rainfall. The analysis also showed significant biases in
individual models depending on the subregion and season
under consideration. These biases appear to be linked to the
model’s failure to resolve convective processes and
topography accurately. The majority of the RCMs used
were consistent with the ground observation in capturing
the dry (wet) conditions associated with the El Nino (La
Nina) events. Statistical analysis conclusively revealed that
the RCMs performance varies over the subregions and
& A. A. Akinsanola
aakinsano2-c@my.cityu.edu.hk
1
School of Energy and Environment, City University of Hong
Kong, Kowloon Tong, Hong Kong SAR
2
Department of Meteorology and Climate Science, Federal
University of Technology Akure, Akure, Nigeria
seasons, implying that no single model is best at all time. In
general, REGCM3 was found to be the most outstanding of
all the RCMs and is therefore recommended for use in
rainfall assessment over West Africa.
Keywords CORDEX-Africa  West African monsoon 
Climatology  Precipitation characteristics
Introduction
Rainfall distribution over West Africa is evidently critical
for many activities such as hydroelectric power generation,
water resources monitoring, rain-fed agriculture, drought,
and flood forecasting (Omotosho and Abiodun 2007; Parry
et al. 2007; Akinsanola and Ogunjobi 2015). Therefore,
variations in rainfall amount have strong and direct
implications on the populace and overall economic growth
of the region. One of the most crucial and dynamic phe-
nomena of the West African climate system is the West
Africa Monsoon (WAM) (Sultan et al. 2011; Akinsanola
et al. 2015), a phenomenon caused by the seasonal reversal
of winds due to differential heating between land and ocean
(Sultan and Janicot 2000; Le Barb́e et al. 2002). The WAM
plays a vital role in producing majority of the annual
rainfall in the region (Omotosho and Abiodun 2007). The
flow dynamics of the WAM is generally characterized by
the intra-annual movement of the intertropical discontinu-
ity (ITD), triggered by the position of the sun, the West
Africa Westerly Jet (WAMJ) which is responsible for the
transport of moisture inland, as well as the strength and
position of the African Easterly Jet (AEJ) and the Tropical
Easterly Jet (TEJ) found at about 600–700 and 200 hPa,
respectively (Jung and Kunstmann 2007; Sylla et al. 2013;
Akinsanola et al. 2015). The combined baroclinic–
123
366 Page 2 of 20
barotropic instabilities of the AEJ are responsible for the
formation of African Easterly Waves (AEW) (Redelsperger
et al. 2002; Burpee 1972). These waves, in turn, trigger the
formation of squall lines and mesoscale convective com-
plexes (Fink and Reiner 2003; Fink et al. 2006). The
above-mentioned components interact in a complex way to
provide the monsoon rainfall (Sultan et al. 2011). Resear-
ches have also shown that interannual and interseasonal
variability occurs in the WAM, and has being linked to the
influence of some global teleconnections and climate
forcing, e.g., El Nino–Southern Oscillation (ENSO), North
Atlantic Oscillation (NAO), Madden–Julian Oscillation
(MJO), and sea surface temperature (Giannini et al. 2003;
Lu and Delworth 2005; Vizy and Cook 2001, 2002).
Presently, there is a limitation in providing skillful and
accurate predictions of WAM variability and its impacts on
the region. Seasonal forecasts based on empirical data are
not always successful (Omotosho and Abiodun 2007). This
is primarily due to gaps in the knowledge of coupled
atmosphere–land–ocean system probably partly arising
from a lack of appropriate ground observation datasets, and
also because of the complex scale interactions between the
atmosphere, biosphere, and hydrosphere, which ultimately
determine the nature of the WAM. (Redelsperger et al.
2006). Providing a solution to the identified gap has
become a major concern to scientists and policy makers
who develop action plans to mitigate and adapt to the
impacts of WAM variability and future climate change.
Climate models are the fundamental and primary tools for
investigating the climate system response to various forc-
ings, making climate predictions from seasonal to decadal
timescales and future climate projection (Akinsanola et al.
2015). However, these dynamic models especially the
global circulation models (GCMs) are not without their
faults, lacking the resolution needed to adequately resolve
ocean wave dynamics, which are important for the correct
representation of El Nino development. Also, the model
physics are incomplete with many processes and feedbacks
on the coupled system being left out. As a result of this
poor resolution in the GCMs, previous researches have
established their unsuitability for weather and impact
studies in regions of complex topography such as West
Africa. Recently, the use of regional climate models
(RCMs) for dynamical downscaling has grown, their res-
olution has increased, process descriptions have been
developed further, new components are added, coordinated
ensemble experiments have become more widespread, and
they also show quite realistic climate signals when com-
pared to observations (Rummukainen 2010; Flato et al.
2013). These models are integrated into a limited domain
and obtain their initial and boundary conditions from glo-
bal climate models or gridded analysis of observations. In
123
Environ Earth Sci (2017) 76:366
this way, a RCM acts as a zooming device to deliver cli-
mate information on regional to local scale. Comprehen-
sive evaluation of regional downscaling is very difficult
especially over West Africa; previous and available studies
often employ different methods, regions, periods, and
observational data for evaluation. Thus, evaluation results
are difficult to generalize. The Coordinated Regional Cli-
mate Downscaling Experiment (CORDEX) is an initiative
that provides a platform for a joint evaluation of model
performance, along with a solid scientific basis for impact
assessments and other uses of downscaled climate infor-
mation (Giorgi et al. 2009).
Published work within CORDEX framework, focusing
on the present climate especially precipitation assessment
over Africa domain, indicates strengths and deficiencies in
the state-of-the-art modeling tools. The first set of the
present-day CORDEX simulations using ERA-Interim
reanalysis at the boundaries has been analyzed in detail
(Nikulin et al. 2012), with focus on precipitation clima-
tology. Results from their study showed a large systematic
bias in RCM-simulated annual mean precipitation over
Africa. Hence, they reported that the multi-model ensemble
mean outperformed individual RCMs. Also, Kim et al.
(2014) reported that the RCMs show difficulties in repli-
cating precipitation in arid and semiarid regions of Africa,
with quasi-systematic wet biases and a poor capability in
reproducing the regional annual cycles. Some other similar
and or related studies (Paeth et al. 2011; Druyan et al.
2010; Diallo et al. 2012, 2013; Sylla et al. 2013; Gbobaniyi
et al. 2013; Klutse et al. 2015; Akinsanola et al.
2015, 2017) also reported that the RCMs performance
varied from one model and region to another. However,
most of the aforementioned previous work focused solely
on the capability of the RCMs to capture the climatological
features and characteristics of the WAM. Presently, there is
still limited information on: (1) the specific seasons of
individual RCMs strength and weakness from a detailed
statistical viewpoint; (2) the basic factors responsible for
the deficiencies in the RCMs; and (3) the capabilities of the
RCMs to simulate the response to El Nino and La Nina
events.
Therefore, providing adequate knowledge and informa-
tion of the capabilities and limitations of these models in
capturing the existing and itemized features is a prerequi-
site before considering climate change simulations over the
region. Hence, this study is aimed at evaluating the ability
of seven CORDEX RCMs to reproduce rainfall charac-
teristics over West Africa; indicate the seasons of indi-
vidual RCMs strength and weakness using a detailed
statistical approach; investigate the factors responsible for
the deficiencies in the RCMs; and also assess the RCMs
ability to simulate response to El Nino and La Nina events.
 Table 1 List of CORDEX RCMs used and their schemes (Nikulin et al. 2012)
DMI-HIRHAM5 ICTP-RegCM3 CLMcom-CCLM4.8 MPI-REMO SMHI-RCA35 UCT-PRECIS UQAM-CRCM5

Institute Danmarks, International Center CLM community (www. Max Plank Institue, Sveriges Meteorologiska University of Cape Universit́e du
running the Meteorologiske for Theoritical clmcommunity.eu ) Germany och Hydrologiska Town, South Africa Qúebec‘ a
Environ Earth Sci (2017) 76:366

model Institut, Danmark Physics, Italy Institut, Sweden Montŕeal,

Canada
Short name HIRHAM REGCM3 CCLM REMO RCA PRECIS CRCM5
Projection Rotated pole 0.44 Mercator 50 km Rotated pole 0.44 Rotated pole 0.44 Rotated pole 0.44 Rotated pole 0.44 Rotated pole 0.44
resolution
Vertical Hybrid 3.1 Sigma/18 Terrain following/3.5 Hybrid/27 Hybrid/40 Hybrid/19 Rotated pole 0.44
coordinates/
levels
Advection Semi-Lagrangian Eulerian Fifth-order upwind Semi-Lagrangian Eulerian Eulerian Semi-Lagrangian
Baldauf and Schulz
(2004)
Time step (s) 600 100 240 240 1200 300 1200
Convective Tiedtke (1989) Grell (1993), Fritsch Tiedtke (1989) Tiedtke (1989) Kain and Fritsch Gregory and Rowntree Kain and Fritsch
scheme and Chappell (1990, 1993) (1990) Gregory and (1990) Kuo
(1980) Allen (1991) (1965)
Radiation Fouquart and Bomel Kiehl et al. (1996) Ritter and Geleyn (1992) Morcrette et al. Savijarvi (1990) Sass Edwards and Slingo Li and Barker
scheme (1980) Mlawer (1986) Giorgetta et al.(1994) (Edwards and Slingo (2005)
et al. (1997) and Wild (1995) 1996)
Turbulence Louis (1979) Holtslag et al. (1990) Herzog et al. (2002), Louis (1979) Cuxart and Bougeault Wilson (1992) Benoit et al.
vertical Buzzi et al. 2011) (Cuxart and Bougeault (1989) Delage
diffusion 2000) (1997)
Cloud Tiedtke (1989), SUBEX Pal et al. Doms et al. (Doms et al. Lohmann and Rasch and Kristjanson Smith (1990) Sundqvist et al.
microphysics Tompkins (2002) (2000) 2007) Baldauf and Roeckner (1996) (1998) (1989)
scheme Schulz (2004)
Land surface Schulz et al. (1998) BATS1E Dickinson TERRA-ML; Doms et al. Hagemann (2002) Samulsson et al. (2006) MOSES2 Essery et al. CLASS 3.5,
scheme Hagemann (2002) et al. (1993) 2007) Rechid et al. (2009) (2003) Verseghy (2000)
Latest Christensen et al. Pal et al. (2007) Rockel et al. (2008), Jacob (2001), Jacob Samuelsson et al. (2011) Jones et al. (2004) Zadra et al. (2008)
reference and (2006) Baldauf et al. (2011) et al. (2007)
comments
Page 3 of 20
366

123
366 Page 4 of 20
Fig. 1 Study domain showing
West Africa topography and the
regions designated as Guinea
Coast, Savanna, and Sahel
zones in the study [Adapted
from Akinsanola et al. 2017]
Data and methodology
In this study, an ensemble of seven RCMs participating in
the CORDEX-Africa program is analyzed and intercom-
pared (details are presented in Table 1). These RCMs were
integrated over Africa continuously for a 20-year period
from 1989 to 2008 at a spatial resolution of 50 km
(*0.44°), employing the ERA-Interim reanalysis dataset
(Dee et al. 2011) at the lateral boundaries and for initial-
ization. The first year of the simulation run was regarded as
spin-up and thus was not included in the analysis. In this
study, the simulations were evaluated over the West Africa
domain (0°–20°N, 20°W–20°E), which exhibits some
localized highlands (Fig. 1). The domain is further subdi-
vided into three climatic zones: Guinea Coast (4°–8°N),
Savannah (8°–11°N), and Sahel (11°–16°N). The Guinea
Coast represents the southern boundary to the Atlantic
Ocean which is characterized by the subhumid climate with
an average annual rainfall of 1250–1500 mm. The Savan-
nah zone is a semiarid zone with an average annual rainfall
of 750–1250 mm, while the Sahel zone covers the northern
boundary of Mauritania, Mali, and Niger. This zone is
essentially characterized by a single rainfall peak within a
short rainy season (June–September), with an average
annual rainfall of about 750 mm.
All the simulation datasets were obtained from the
Climate Systems Analysis Group (CSAG) of the University
of Cape Town, Cape Town, South Africa. Rainfall data
from ground observation, Climate Research Unit (CRU),
and Tropical Rainfall Measuring Mission (TRMM) were
123
Environ Earth Sci (2017) 76:366
used to compare and validate the RCMs. The ground
observation data were obtained from the African rainfall
database of the Institute of Geophysics and Meteorology of
the University of Cologne, Germany, for 81 meteorological
stations in West Africa covering the period 1990–2008,
detail description in (Akinsanola et al. 2015, 2016); the
Climate Research Unit (CRU) Time Series (TS) at the
University of East Anglia, CRU TS 3.24 (Mitchell and
Jones 2005; Harris et al. 2014) is from 1990 to 2008 at 0.5°
by 0.5° spatial resolution while the TRMM rainfall was
obtained from the Tropical Rainfall Measurement Mission
Multi-Platform Analysis (TMPA 3B42V7) (Kummerow
et al. 2001; Huffman et al. 2001) at 0.25° by 0.25° spatial
resolutions from 1998 to 2008. The 850 hPa wind and
specific humidity data used were retrieved from the ERA-
Interim reanalysis, a product from European Centre for
Medium-range Weather Forecast (ECMWF) at 0.5° by 0.5°
spatial resolution from 1990 to 2008.
Since all the products used in this study had different
spatial and temporal resolutions, they were first re-gridded
to a spatial resolution of *0.44° (50 km) using bilinear
interpolation as in Nikulin et al. (2012). Also, to address
the uniformity in the temporal span of all the dataset, the
rainfall data were aggregated into monthly means over the
period of 1990–2008. It should be noted that the averaging
periods of the TRMM product is necessarily different due
to the limited availability periods of the data. However, we
assume that this period (i.e., 1998–2008) is informative
enough to assess systematic biases, capture year-to-year
and seasonal variability, and that the observed differences
Environ Earth Sci (2017) 76:366
Fig. 2 Spatial pattern of mean monthly DJF rainfall (mm/month) for a CRU, b TRMM, c PRECIS, d RCA, e REMO, f CCLM, g CRCM5,
h HIRHAM, i REGCM3 from 1990 to 2008
between the products are dominated by the respective
measurement technique rather than long-term climate
variability (Hijmans et al. 2005; Akinsanola et al. 2016).
Four seasons were selected to compare and study the
rainfall estimates in detail: the dry season (December–Fe-
bruary), the pre-rainy season (March–May), the rainy
season (June–August), and the post-rainy season
(September–November). A comparison was performed for
the three main subregions over West Africa. For each
subregion and season, the comparison includes a climato-
logical and statistical description. The climatological
description is presented by showing the spatial precipita-
tion pattern, latitude–time cross section, interannual and
interseasonal variability, the spatial distribution of specific
humidity and wind, and lastly, the RCMs depiction of El
Nino and La Nina events. For the statistical description, the
rainfall simulations of the CORDEX RCMs were assessed
using Taylor diagrams (Taylor 2001), which graphically
Page 5 of 20 366
synthesize the degree of correspondence between RCMs
and the ground observation in terms of the phase and
amplitude of their evolution, measured by Pearson corre-
lation coefficients, the centered root mean square error
(RMSE), mean bias error (MBE), and standard deviation
(SD). This diagram has been widely used to evaluate the
multiple aspects of complex models and gauging the rela-
tive skill of many different models (e.g., Kalognomou et al.
2013; Akinsanola et al. 2015).
Furthermore, in order to access extreme events of El
Nino and La Nina, the standardized rainfall index was
calculated for all the composite years of the two events (the
events were sourced from http://ggweather.com/enso/oni.
htm). This was done using Eq. 1.
xx
u ¼ ð1Þ
r
where u represents the standardized rainfall index, x is the
actual value of rainfall, x is the long-term mean value of
123
366 Page 6 of 20
Fig. 3 Spatial pattern of mean monthly MAM rainfall (mm/month) for a CRU, b TRMM, c PRECIS, d RCA, e REMO, f CCLM, g CRCM5,
h HIRHAM, i REGCM3 from 1990 to 2008
rainfall, r is the standard deviation. The dry and wet years
have been classified following Mckee et al. (1993).
Results and discussion
Mean climatology
It is very important to assess the abilities and degree at
which the RCMs capture the spatial pattern of rainfall
over West Africa. Hence, the spatial distribution of the
mean December–January–February (DJF) rainfall from
(1990 to 2008) over West Africa is shown in Fig. 2a–i for
CRU, TRMM, and each of the RCMs used. It was
observed that larger portion of West Africa is having
rainfall below 20 mm especially north of latitude 8°N.
However, rainfall was found to be highest in the eastern
part of the Guinea Coast (about longitude 10°E). The
123
Environ Earth Sci (2017) 76:366
RCMs are consistent in capturing the decreasing rainfall
from south to north during this period. The pattern
obtained for the months of March–April–May (MAM)
presented in Fig. 3 is similar to that of DJF. Rainfall
amount over the larger part of West Africa is found to
have increased reaching the Savannah region (about
11°N), although little or no rainfall is still highly pre-
dominant over the Sahel. The zone of highest rainfall is
observed to be around the coastal part extending from
Liberia toward the southeastern part of Nigeria. This
observed region of low and high rainfall amount is con-
sistent in all the datasets, but the amount of rainfall
attributed varies to a very large extent from one model to
another. Furthermore, the summer June–July–August
(JJA) pattern of precipitation as shown in Fig. 4 reveals
that rainfall is observed over the entire West Africa
domain with the Guinea Coast having the highest rainfall
amount. In addition, the zones of maximum rainfall are
Environ Earth Sci (2017) 76:366
Fig. 4 Spatial pattern of mean monthly JJA rainfall (mm/month) for a CRU, b TRMM, c PRECIS, d RCA, e REMO, f CCLM, g CRCM5,
h HIRHAM, i REGCM3 from 1990 to 2008
localized on the major West Africa highlands earlier
indicated in Fig. 1. The orography-related rainfall maxi-
mum found over the Cameroon Mountains extends from
northeast to southwest, with an anvil-like shape encom-
passing Nigeria and Lake Chad to the left and right,
respectively. These observations are well captured by the
RCMs. Figure 5 shows the same pattern as Fig. 2 but for
the months of September–October–November (SON). The
amount of rainfall has reduced drastically compared to
that of JJA, although CRCM5 and HIRHAM exhibited a
large bias. It can be concluded that rainfall decreases
northward as far as the southern fringe of the Sahel region
throughout the season. Also, orography plays an impor-
tant role in West Africa rainfall patterns as reported by
Jenkins et al. (2002); Jones et al. (2011); and Akinsanola
et al. (2015, 2017). The overestimations of rainfall
amount exhibited by some of the models over the
Page 7 of 20 366
highlands may be linked to the model’s failure to resolve
topography accurately.
Mean annual cycle and Interannual variation
The latitude–time cross section of the mean monthly
rainfall over West Africa averaged between longitude
10°W and 10°E is shown in Fig. 6. Here, the ability of the
RCMs to capture the three distinctive phases of the WAM
(i.e., the onset, the high rain period and the southward
retreat) is evaluated. The TRMM data exhibit onset of
rainfall around late March and early June, with a pro-
gressive extension of the rain belt from the Guinea Coast to
about latitude 8°N. Also observed, is a sharp discontinuity
known as the monsoon jump as similarly reported in Sultan
and Janicot (2000, 2003) and Akinsanola et al. (2015). This
jump occurs in the rain band between late June and early
123
366 Page 8 of 20
Fig. 5 Spatial pattern of mean monthly SON rainfall (mm/month) for a CRU, b TRMM, c PRECIS, d RCA, e REMO, f CCLM, g CRCM5,
h HIRHAM, i REGCM3 from 1990 to 2008
July and is characterized by a northward shift of the
monsoon rainfall front to about 10°N. This brings high
rainfall amounts into the Sahel accompanied by a sudden
cessation in rainfall amount along the Guinea Coast
(Gbobaniyi et al. 2013; Akinsanola et al. 2015). Further-
more, around September, the rain band experiences a
southward shift toward the Guinea Coast accompanied by a
decrease in rainfall amount in the Sahel. The RCMs cap-
tured the three distinct phases of the mean annual cycle of
the WAM fairly. However, a number of differences were
found among the RCMs with regard to the magnitude and
spatial extent of the WAM features. For instance, RCA and
CRCM5 fail to reproduce the monsoon jump distinctively
while CCLM, HIRHAM and REMO largely overestimate
the intensity of the pre- and post-monsoon rainfall. It is
123
Environ Earth Sci (2017) 76:366
worth mentioning that most of the deviations and dis-
crepancies occur with respect to the peak of monsoon
rainfall over the Sahel. Earlier studies have concluded that
divergences in the RCMs annual cycles arise mostly from
their different inabilities in simulating the appearance and
development of the main features responsible for triggering
and maintaining the WAM rainfall. Among them, we have
the monsoon flow, AEJ, TEJ, and AEWs (Diallo et al.
2012; Sylla et al. 2013). Therefore, the RCM analyses here
may exhibit quite different sensitivities in terms of their
response to the intensity of the WAM elements.
The zonal average of the monthly rainfall was evaluated
using the ground observation data (81 stations) as reference
over the three homogeneous subregions and the entire West
Africa domain from 1990 to 2008 as shown in Fig. 7. This
Environ Earth Sci (2017) 76:366 Page 9 of 20 366

Fig. 6 Latitude–time cross section of mean monthly rainfall (mm/month) for a CRU, b TRMM, c PRECIS, d RCA, e REMO, f CCLM,
g CRCM5, h HIRHAM, i REGCM from 1990 to 2008 averaged over longitude 10 W to 10E

helps to better identify rainfall minima and peaks, and thus
to further gain insights about the capability of the RCMs in
capturing phases and amplitudes during the course of the
year in specific homogenous zones. Over the Guinea Coast,
the ground observation data exhibited two peaks of rainfall,
a primary maximum in June, and a secondary in Septem-
ber. Also present is a relative mid-summer low (minimum)
in August known as the little dry season (LDS); this occurs
as the monsoon rain band seasonally migrates in the north–
south direction in the Guinea Coast. A slight difference was
observed in the amount of rainfall among the RCMs, with
REMO showing the largest magnitude and REGCM3 the
smallest, thus, respectively, overestimating and underesti-
mating the amount of both peaks. It is worth emphasizing
that while all the RCMs capture the timing of the mid-
summer break, the majority of them simulate an early
primary peak in May, but only a few of them shift the
secondary maximum to September. These observed dif-
ferences in the RCM-simulated rainfall may be attributed
to their different sensitivities and responses to the pre-
scribed sea surface temperature as reported by Thorncroft
et al. (2011). In the Savannah, the length of rainy season

123
366 Page 10 of 20 Environ Earth Sci (2017) 76:366

(a) (b)
400 CCLM CRCM5 HIRHAM 350
REGCM3 PRECIS RCA
Rainfall (mm/month) 350 REMO Ground observaon 300

Rainfall (mm/month)
300 250
250
200
200
150
150
100
100
50
50
0
0 JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC
JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC
Month
Month
(c) (d)
250 250
Rainfall (mm/month)

200 200

Rainfall (mm/month)
150 150

100 100

50 50

0 0
JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC
Month Month

Fig. 7 Zonal variation of mean monthly rainfall (mm/month) for the ground observation and CORDEX RCMs data over a Guinea Coast,
b Savannah, c Sahel, and d West Africa from 1990 to 2008

(a) (b)
100
100
80
80
60
60
40
% difference

40
% difference

20
20
0
0 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008
1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 -20
-20
-40
-40
Year Year
-60
-60
-80
-80 CCLM CRCM5 HIRHAM REGCM3
PRECIS RCA REMO -100
-100

(c) (d)
100
100
80
80
60
60
40
% difference

40
% difference

20
20
0
0 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008
1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 -20
-20
-40
-40 Year
Year -60
-60
-80
-80
-100
-100

Fig. 8 Inter annual percentage difference of the CORDEX RCMs from ground observation over a Guinea Coast b Savannah c Sahel and
d whole West Africa from 1990 to 2008

123
Environ Earth Sci (2017) 76:366 Page 11 of 20 366

Fig. 9 Standardized rainfall (a)

index during JJA for ground
observation and CORDEX 7 1991/1992 1997/1998 2002/2003
RCMs associated with a El
Nino events of 1991/1992, 5
1997/1998 and 2002/2003 and
b La Nina events of 1998/1999, 3
1999/2000, 2000/2001 over
West Africa 1

SPI
-1 Ground CCLM CRCM5 HIRHAM REGCM3 PRECIS RCA REMO
observaon
-3

-5

-7

(b)
5
1998/1999 1999/2000 2000/2001
4

2
SPI

0
Ground CCLM CRCM5 HIRHAM REGCM3 PRECIS RCA REMO
observaon
-1

-2

extends from May to September, rainfall over this region
exhibits a unimodal pattern, having its peak in the month of
August. Majority of the models captured the observed
rainfall distribution but underestimated the rainfall amount
during the summer month. HIRHAM was observed to have
overestimated rainfall while PRECIS captured two peaks
over this zone. In the Sahel, the length of the rainy season
is less than four months between July and September, with
August having the highest rainfall amount. All the RCMs
reproduce the observed rainfall peak in August except
CRCM5. REGCM3, HIRHAM, PRECIS, and REMO tend
to overestimate its rainfall amount while CRCM5 and RCA
considerably underestimated the rainfall amount. In addi-
tion, HIRHAM simulates an early peak in July while
CCLM shifted the rainfall peak to September. Over the
entire West Africa domain, the highest rainfall amount was
observed between July and September with a great con-
sistency in all the RCMs.
The percentage difference of the RCMs from the ground
observation based on the interannual variation is shown in
Fig. 8. The majority of the models depict a large positive
difference in the subregions with relatively low deviations
over the larger domain of West Africa (about 20%). These
indicate that the RCMs have the best performance over a
large domain owing to their spatial resolution. The findings
presented here are in line with the work of Browne and
Sylla (2012), Akinsanola et al. (2017).
Depiction of El Nino and La Nina events
The influence of El Nino–Southern Oscillation (ENSO) on
rainfall over West Africa has been evaluated in the previ-
ous study (see review in Moron and Ward 1998). Nichol-
son and Entekhabi (1986), Ropelewski and Halpert
(1987, 1989), and Nicholson and Kim (1997) suggest that
ENSO influence is minimal, while some others (e.g.,
Hastenrath 1990; Bhatt 1989; Moron and Ward 1998)
suggest that ENSO tends to reduce rainfall in the Sahel.
Hence, this study examined rainfall anomalies over West
Africa during the summer monsoon months of June–July–
August (JJA) associated with El Nino and La Nina events
as shown in Fig. 9a and b, using the ground observation
data of 81 stations as the reference. Negative anomalies
were observed during the El Nino years which implies dry
conditions. All the RCMs captured these feature; however,
it is worth mentioning that RCA overestimated the event

123
366 Page 12 of 20
Fig. 10 Spatial distribution of mean JJA 850 hPa specific humidity
(g/kg) and wind vector (m/s) for a ERA-Interim, b CCLM, c CRCM5,
d HIRHAM, e REGCM3, f RCA, and g REMO from 1900 to 2008
while REGCM3, REMO failed to capture the El Nino
events of 2002/2003; furthermore, HIRHAM failed to
capture the event in all the years (see Fig. 9a). The failure
of HIRHAM in capturing the dry conditions associated
with El Nino events may be related to rainfall overesti-
mation earlier observed in the model results during summer
month. In the La Nina events (see Fig. 9b), positive
anomalies were observed implying a wet condition,
majority of the RCMs were consistent with the ground
observation, although, little underestimation was observed
in CCLM and CRCM5. Generally, the two climate events
are well captured by the RCMs.
Wind and humidity field
The performance of the model over the West Africa
domain requires further examination of the dynamics of
123
Environ Earth Sci (2017) 76:366
over West Africa (Note: PRECIS data are not available; CRCM5
specific humidity data are also not available)
the monsoon. Therefore, the wind field and specific
humidity distribution at 850 hPa during the summer
months of JJA from 1990 to 2008 was investigated and
is presented in Fig. 10. The ERA-Interim depicts a nar-
row band of westerlies located at about 4°N and 8°N,
these westerlies propagating from the Atlantic Ocean and
extend eastward as far as 20°E, with a maximum west-
erly flow of 3.33 m/s centered at approximately 12°E
over the Cameroon mountains. Similarly, a weak west-
erly (1 m/s) was observed over the continent and it
extends northward to 18°N, overlain by easterlies over
the Saharan region. Between latitude 15° and 20°N, two
peaks of easterlies are embedded. The first is located
between 20° and 10°W with a maximum wind speed of
about 5.7 m/s. The second is located between 10° and
20°E and is comparable in magnitude to the first east-
erlies. The RCM-simulated wind speed is stronger, and
Environ Earth Sci (2017) 76:366 Page 13 of 20 366

Fig. 11 Taylor diagrams for area-averaged monthly rainfall over The radial coordinate shows the standard deviation. The azimuthal
Guinea Coast for a DJF, b MAM, c JJA, d SON. The reference axis shows the correlation between the RCMs and the reference data.
(ground observation) data are shown by the green square along the The centered root mean square error is indicated by the green
horizontal axis. The individual RCMs are shown by the solid circles. semicircles about the reference point

Fig. 12 Taylor diagrams for area-averaged monthly rainfall over coordinate shows the standard deviation. The azimuthal axis shows
Savannah for a DJF, b MAM, c JJA, d SON. The reference (ground the correlation between the RCMs and the reference data. The
observation) data are shown by the green square along the horizontal centered root mean square error is indicated by the green semicircles
axis. The individual RCMs are shown by the solid circles. The radial about the reference point

123
366 Page 14 of 20 Environ Earth Sci (2017) 76:366

Fig. 13 Taylor diagrams for area-averaged monthly rainfall over coordinate shows the standard deviation. The azimuthal axis shows
Sahel for a DJF, b MAM, c JJA, d SON. The reference (ground the correlation between the RCMs and the reference data. The
observation) data are shown by the green square along the horizontal centered root mean square error is indicated by the green semicircles
axis. The individual RCMs are shown by the solid circles. The radial about the reference point

Fig. 14 Taylor diagrams for area-averaged monthly rainfall over the The radial coordinate shows the standard deviation. The azimuthal
whole West Africa for a DJF, b MAM, c JJA, d SON. The reference axis shows the correlation between the RCMs and the reference data.
(ground observation) data are shown by the green square along the The centered root mean square error is indicated by the green
horizontal axis. The individual RCMs are shown by the solid circles. semicircles about the reference point

123
Environ Earth Sci (2017) 76:366 Page 15 of 20 366

Table 2 Detailed output of the statistical analysis between the RCMs Table 2 continued
and the ground observation over Guinea Coast and Savannah
Model MBE RMSE r Remarks
Model MBE RMSE r Remarks
Savannah (JJA)
Guinea Coast (DJF) RCA 217.15 32.69 0.91 Suitable
RCA 1.38 2.46 0.81 Suitable PRECIS 243.67 58.29 0.84 Suitable
PRECIS 81.06 81.14 -0.58 Not Suitable CRCM5 239.91 45.69 0.97 Suitable
CRCM5 5.43 10.37 -0.19 Not Suitable REMO -8.29 48.93 0.45 Not Suitable
REMO 20.57 2.28 0.96 Suitable REGCM3 0.69 24.70 0.92 Most Suitable
REGCM3 0.81 1.24 0.75 Most Suitable CCLM -105.80 125.73 -1.00 Not Suitable
CCLM 1.29 6.10 0.04 Not Suitable HIRHAM 60.78 66.26 0.79 Suitable
HIRHAM 11.08 13.62 0.88 Suitable Savannah (SON)
Guinea Coast (MAM) RCA 5.82 6.93 1.00 Most Suitable
RCA 45.54 49.81 0.98 Suitable PRECIS 7.18 20.48 0.99 Suitable
PRECIS 30.34 41.53 0.24 Not Suitable CRCM5 85.74 93.30 0.92 Suitable
CRCM5 214.27 16.56 0.98 Suitable REMO 37.81 47.78 0.99 Suitable
REMO 35.52 49.48 0.99 Suitable REGCM3 11.46 24.54 1.00 Suitable
REGCM3 4.26 6.45 1.00 Most Suitable CCLM 24.72 62.97 0.77 Suitable
CCLM 24.12 29.83 0.65 Suitable HIRHAM 37.17 44.53 0.98 Suitable
HIRHAM 68.84 73.19 0.82 Suitable
Note best performing models are in bold
Guinea Coast (JJA)
RCA 223.25 29.93 0.66 Suitable
PRECIS -84.99 89.09 -0.91 Not Suitable
CRCM5 -23.46 33.07 0.37 Not Suitable
REMO 4.32 27.06 0.04 Not Suitable Table 3 Detailed output of the statistical analysis between the RCMs
REGCM3 25.26 6.37 0.96 Most Suitable and the ground observation over Sahel and West Africa
CCLM -39.57 55.65 -0.99 Not Suitable Model MBE RMSE r REMARKS
HIRHAM 62.85 66.76 0.82 Suitable
Guinea Coast (SON) Sahel (DJF)
RCA 6.03 9.59 1.00 Most Suitable RCA 2.79 2.91 0.64 Suitable
PRECIS 2.68 57.15 -0.99 Not Suitable PRECIS 3.99 4.10 0.09 Not Suitable
CRCM5 37.64 50.56 0.86 Suitable CRCM5 38.81 38.81 -0.98 Not Suitable
REMO 35.50 39.48 0.97 Suitable REMO 1.01 1.05 -0.51 Not Suitable
REGCM3 10.28 15.95 0.98 Suitable REGCM3 1.73 1.77 0.37 Not Suitable
CCLM 46.08 57.44 0.85 Suitable CCLM 3.58 4.52 -0.97 Not Suitable
HIRHAM 51.12 54.73 0.99 Suitable HIRHAM 4.53 4.99 0.81 Most Suitable

Savannah (DJF) Sahel (MAM)

RCA 4.37 5.40 0.92 Most Suitable RCA 67.83 70.50 0.97 Suitable

PRECIS 6.52 9.47 0.79 Suitable PRECIS 52.41 53.99 0.96 Suitable

CRCM5 5.28 10.79 -0.91 Not Suitable CRCM5 216.89 20.92 0.97 Suitable

REMO 22.38 2.45 0.69 Suitable REMO 17.51 26.29 1.00 Suitable

REGCM3 -0.90 1.44 -0.41 Not Suitable REGCM3 6.31 9.38 1.00 Most Suitable
CCLM -2.39 2.56 -0.42 Not Suitable CCLM 218.88 25.05 0.98 Suitable
HIRHAM 9.90 13.34 0.96 Suitable HIRHAM 68.78 73.06 1.00 Suitable
Savannah (MAM) Sahel (JJA)
RCA 45.54 49.81 0.98 Suitable RCA 221.10 36.20 0.94 Suitable
PRECIS 30.34 41.53 0.24 Not Suitable PRECIS 29.35 38.78 0.92 Suitable
CRCM5 214.27 16.56 0.98 Suitable CRCM5 -79.21 96.36 -0.98 Not Suitable
REMO 35.52 49.48 0.99 Suitable REMO 24.79 28.41 0.97 Most Suitable
REGCM3 4.26 6.45 1.00 Most Suitable REGCM3 26.81 30.07 0.97 Suitable
CCLM 24.12 29.83 0.78 Suitable CCLM 218.94 33.73 0.95 Suitable

HIRHAM 68.84 73.19 0.81 Suitable HIRHAM 48.05 53.74 0.95 Suitable

123
366 Page 16 of 20 Environ Earth Sci (2017) 76:366

Table 3 continued flow at 12°E, but locates it farther north along Lake
Model MBE RMSE r REMARKS
Chad at 10°N, with a dominance of cyclonic circulation
over Mauritania. The observed circulation may imply
Sahel (SON) high moisture convergence and upward motion owing to
RCA 5.29 7.91 0.99 Most Suitable the presence of high humidity of about 10–15 g/kg, and
PRECIS 23.22 25.15 0.99 Suitable these may in turn result into heavy rainfall. These
CRCM5 0.98 41.95 1.00 Suitable observations may be the reason for overestimation earlier
REMO 35.94 41.68 0.99 Suitable observed in the RCMs. Ultimately, the RCMs simulate
REGCM3 14.25 22.89 1.00 Suitable monsoon that penetrates more northward to about 16°N,
CCLM 31.26 34.27 0.99 Suitable covering the lower half of the Sahel. The difference in
HIRHAM 35.62 40.37 0.99 Suitable the northward extent of the monsoon between ERA-In-
West Africa (DJF) terim and RCMs may be due to the differences in the
RCA 1.38 2.46 0.81 Suitable strength of the simulated westerlies as indicated earlier,
PRECIS 81.06 81.14 -0.58 Not Suitable in particular, the westerly jet. In fact, studies have shown
CRCM5 5.43 10.37 -0.19 Not Suitable that the low-level westerly jet is an important agent for
REMO 20.57 2.28 0.96 Suitable moisture transporting from the eastern tropical Atlantic
REGCM3 0.81 1.24 0.75 Most Suitable Ocean into continental of West Africa during boreal
CCLM 1.29 6.10 0.04 Not Suitable summer (Patricola and Cook, 2007; Pu and Cook
HIRHAM 11.08 13.62 0.88 Suitable 2010, 2011).
West Africa (MAM)
RCA 45.54 49.81 0.98 Suitable Statistical evaluation
PRECIS 30.34 41.53 0.24 Not Suitable
CRCM5 214.27 16.56 0.98 Suitable
Taylors diagram (Taylor 2001) was employed to assess the
REMO 35.52 49.48 0.99 Suitable
models’ ability in skillfully simulating rainfall over the
three homogeneous subregions and the entire West Africa
REGCM3 4.26 6.45 1.00 Most Suitable
domain as presented in Figs. 11, 12, 13, 14 and later
CCLM 24.12 29.83 0.72 Suitable
summarized in Tables 2 and 3. Seasonal means of
HIRHAM 68.84 73.19 0.69 Suitable
December–February (DJF), March–May (MAM), June–
West Africa (JJA)
August (JJA), and September–November (SON) were used
RCA 223.25 29.93 0.66 Suitable
for the computation. The results shown are based on the
PRECIS 84.99 -0.91 -0.60 Not Suitable
interannual variation of seasonal mean rainfall for the
CRCM5 -23.46 33.07 0.37 Not Suitable
period 1990–2008. Each RCM were compared against the
REMO 4.32 27.06 0.04 Not Suitable
ground observation data using mean bias error (MBE),
REGCM3 25.26 6.37 1.00 Most Suitable
standard deviation (SD), root mean square error (RMSE)
CCLM -39.57 55.65 -0.99 Not Suitable
and Pearson’s correlation coefficient (r). Over the Guinea
HIRHAM 62.85 66.76 0.76 Suitable
Coast (Fig. 11), the RCMs exhibits a remarkable results
West Africa (SON) r [ 0.7 and having a standard deviation in the vicinity of
RCA 1.93 4.27 0.96 Most Suitable the ground observation in most of the seasons. PRECIS
PRECIS -1.42 58.90 -0.89 Not Suitable failed to reproduce the magnitude of the interannual vari-
CRCM5 33.54 47.62 0.89 Suitable ability in all the seasons while CCLM failed to reproduce
REMO 31.40 34.12 0.98 Suitable the observed rainfall during the DJF and JJA (see Table 2).
REGCM3 6.17 13.81 0.99 Suitable In the Savannah (Fig. 12), most of the RCMs were able to
CCLM 41.98 54.19 0.88 Suitable reproduce the observed rainfall in all the seasons
HIRHAM 47.02 49.72 0.99 Suitable (r [ 0.69). Although, CCLM exhibited a poor performance
Note best performing models are in bold during the DJF and JJA seasons while CRCM5 and
REGCM3 also failed to reproduce the observed rainfall
variability in the DJF seasons. In the Sahel and entire West
Africa domain (Figs. 13 and 14 and Table 3), majority of
the westerly jet spans more latitudes than the ERA-In- the RCMs exhibited a remarkable performance in all the
terim. The observed westerlies in the RCMs extend seasons, except in the Sahel during the DJF season where
zonally between latitude 4° and 12°N with local westerly majority of the model exhibited poor performance. It is
maxima of 6.53 m/s in situ at approximately 12°N. Also, evident that the RCMs performance varies from one sub-
the majority of the RCMs simulates maximum westerly region and season to another, implying that no single

123
Environ Earth Sci (2017) 76:366
model is best at all time. However, based on individual
model performances, REGCM3 consistently outperformed
all the other models over the three subregions and the entire
West Africa domain in all seasons.
Conclusion
This study has evaluated the ability of seven CORDEX
regional climate models (DMI-HIRHAM5, ICTP-REGCM3,
CLMcom-CCLM4.8, MPI-REMO, SMHI-RCA35, UCT-
PRECIS, and UQAM_CRCM5) in reproducing the rainfall
characteristics over West Africa. These models have been
evaluated against TRMM and CRU rainfall products, while
the ground observation rainfall data from 81 ground
observing stations over West Africa have been further
employed for comparison and validation. In simulating the
spatial rainfall distribution, the overall performance of the
RCMs is good, particularly to the north of the study domain.
Other aspects of the simulated rainfall that are of practical
significance such as the three distinctive stages of the West
African monsoon is well captured by some of the RCMs,
although with noticeable bias. The significant bias found in
some of the RCMs ranges from failure to reproduce the
monsoon jump observed in RCA and CRCM5 to consider-
able overestimation of rainfall amount during pre- and the
post-monsoon periods observed in CCLM, HIRHAM, and
REMO. Furthermore, the mean annual cycle of rainfall was
fairly replicated by the RCMs; this includes the single
(double-peaked) rainfall over the Sahel (Guinea Coast).
Also, the RCMs reproduce considerably the interannual
rainfall variation; the subregions had higher positive bias
reaching up to 70% while the larger domain of West Africa
exhibited the lowest bias of about 20%. These results imply
that the RCMs had the best performance over a larger
domain than over the subregions, thus emphasizing the
importance of subdividing a large domain such as West
Africa into different homogeneous subregions for evaluation
analyses. Despite the overall satisfactory performance of the
RCMs, it should be emphasized that individual models
exhibited noticeable bias, which largely varied in magnitude
from one season and subregion to another.
Ultimately, the RCMs largely reproduced the corre-
sponding negative (positive) rainfall anomalies observed
during the El Nino (La Nina) years, although RCA over-
estimated the El Nino events for all the years considered
while REGCM3, REMO failed to capture the El Nino
events of 2002/2003. The HIRHAM, however, failed to
capture all the El Nino events considered. Compared to the
ERA-Interim, the RCMs simulated a stronger wind field at
850 hPa and had wider spatial extent of strong westerly
wind. In addition, an intense cyclonic circulation and high
humidity were observed around Mauritania (11°–17°N, 0°–
Page 17 of 20 366
15°W) in the majority of the models. These observations
may be the reason for the earlier reported overestimation in
rainfall amount. Lastly, the result of the statistics indicated
that the model’s performance varied from one subregion to
another, as well as from season to season. However, on the
basis of individual model performance REGCM3 consis-
tently outperformed all the other models as it captured
almost all the basic West Africa rainfall characteristics.
Therefore, this study suggests its use for impact studies of
climate variability and change over the region.
Acknowledgements The authors appreciate WCRP and START for
setting up and funding the CORDEX-Africa analysis initiative and the
University of Cape Town for leading the training and analysis program.
We are very grateful to regional downscaling groups who kindly shared
the downscaled data used in this analysis. We are grateful to the services
that have operated the TRMM and CRU and also the African rainfall
database of the Institute of Geophysics and Meteorology, University of
Cologne, Germany, for the provision of the ground observation data.
The efforts of Mr. Abolude Akintayo Temiloluwa and the two anony-
mous reviewers are also acknowledged.
References
Akinsanola AA, Ogunjobi KO (2015) Recent Homogeneity Analysis
and Long Term Spatio-temporal Rainfall Trends in Nigeria.
Theor Appl Climatol. doi:10.1007/s00704-015-1701-x
Akinsanola AA, Ogunjobi KO, Gbode IE, Ajayi VO (2015) Assessing
the capabilities of three regional climate models over CORDEX
Africa in simulating West African summer monsoon precipita-
tion. Adv Meteorol 2015. doi:10.1155/2015/935431
Akinsanola AA, Ogunjobi KO, Ajayi VO, Adefisan EA, Omotosho
JA, Sanogo S (2016) Meteorol Atmos Phys. doi:10.1007/s00703-
016-0493-6
Akinsanola AA, Ajayi VO, Adejare AT, Adeyeri OE, Gbode IE,
Ogunjobi KO, Nikulin G, Abolude AT (2017) Evaluation of
rainfall simulations over West Africa in dynamically downscaled
CMIP5 global circulation models. Theor Appl Climatol. doi:10.
1007/s00704-017-2087-8
Baldauf M, Schulz JP (2004) Prognostic precipitation in the Lokal
Modell (LM) of DWD. COSMO Newsletter, No. 4, DWD,
Offenbach, Germany, 177–180
Baldauf M, Seifert A, Forstner J, Majewski D, Raschendorfer M,
Reinhardt T (2011) Operational convective-scale numerical
weather prediction with the COSMO model: description and
sensitivities. Mon Weather Rev 139:3887–3905
Benoit R, Cote J, Mailhot J (1989) Inclusion of a TKE boundary layer
parameterization in the Canadian regional finite-element model.
Mon Weather Rev 117:1726–1750
Bhatt U (1989) Circulation regimes of rainfall anomalies in the
Africa-South Asian monsoon belt. J Climate 2:1133–1144
Browne NAK, Sylla MB (2012) Regional climate model sensitivity to
domain size for the simulation of the West African monsoon
rainfall. Int J Geophys. doi:10.1155/2012/625831
Burpee RW (1972) The origin and structure of easterly waves in the
lower troposphere of North Africa. J Atmos Sci 29:77–90
Buzzi M, Rotach MW, Raschendorfer M, Holtslag AAM (2011)
Evaluation of the COSMO-SC turbulence scheme in a shear-
driven stable boundary layer. Meteor Z 20:335–350
Christensen OB, Drews M, Christensen JH (2006) The HIRHAM
regional climate model version 5. DMI Tech. Rep. 06–17, pp 22
123
366 Page 18 of 20
Cuxart J, Bougeault Redelsperger JL (2000) A turbulence scheme al-
lowing for mesoscale and large-eddy simulations. Q J R
Meteorol Soc 126:1–30
Dee DP et al (2011) The ERA-Interim reanalysis: configuration and
performance of the data assimilation system. Q J R Meteorol Soc
137:553–597
Delage Y (1997) Parameterising sub-grid scale vertical transport in
atmospheric models under statically stable conditions. Bound
Layer Meteorol 82:23–48
Diallo I, Sylla MB, Camara M, Gaye AT (2012) Interannual
variability of rainfall and circulation features over the Sahel
based on multiple regional climate models simulations. Appl
Climatol, Theor. doi:10.1007/s00704-012-0791-y
Diallo I, Sylla MB, Camara M (2013) Gaye AT (2013) Interannual
variability of rainfall and circulation features over the Sahel
based on multiple regional climate models simulations. Theor
Appl Climatol 113(1–2):351–362
Dickinson RE, Henderson-Sellers A, Kennedy PJ (1993) Biosphere-
atmosphere transfer scheme (BATS) version 1E as coupled to the
NCAR community climate model. NCAR Technical Note
NCAR/TN-387?STR. doi:10.5065/D67W6959
Doms G, Forstner J, Heise E, Herzog HJ, Raschendorfer M, Schrodin
R, Reinhardt T, Vogel G (2007) A description of the nonhydro-
static regional model LM (version 3.20).Part II: Physical
parameterization. Consortium for Small Scale Modelling Rep,
p 161
Druyan LM, Feng J, Cook KH, Xue Y, Fulakeza M, Hagos SM,
Konaré A, Moufouma-Okia W, Rowell DP, Vizy EK, Ibrah SS
(2010) The WAMME regional model intercomparison study.
Clim Dyn 35(1):175–192. doi:10.1007/s00382-009-0676-7
Edwards JM, Slingo A (1996) Studies with a flexible new radiation
code. I: choosing a configuration for a large-scale model. Q J R
Meteorol Soc 122:689–719
Essery RLH, Best MJ, Betts RA, Cox PM (2003) Explicit represen-
tation of sub grid heterogeneity in a GCM land surface scheme.
J Hydrometeor 4:530–543
Fink AH, Reiner A (2003) Spatiotemporal variability of the relation
between African easterly waves and West African squall lines in
1998 and 1999. J Geophys Res 108(D11):4332. doi:10.1029/
2002JD002816
Fink AH, Vincent DG, Ermert V (2006) Rainfall types in the West
African Sudanian zone during the summer monsoon 2002. Mon
Weather Rev 134:2143–2164
Flato G, Marotzke J, Abiodun B, Braconnot P, Chou S, Collins W,
Cox P, Driouech F, Emori S, Eyring V, Forest C, Glecker P,
Guilyardi E, Jacob C, Kattsov V, Reason C, Rummukainen M
(2013) Evaluation of Climate Models, in: Climate Change: The
Physical Science Basis. Contribution of Working Group I to the
Fifth Assessment Report of the Intergovernmental Panel on
Climate Change, Cambridge University Press, Cambridge,
United Kingdom and New York, NY, USA
Fouquart Y, Bonnel B (1980) Computations of solar heating of the
earth’s atmosphere: a new parameterization. Beitr Phys Atmos
53:35–62
Fritsch JM, Chappell CF (1980) Numerical prediction of convectively
driven mesoscale pressure systems. Part I: convective parame-
terization. J Atmos Sci 37:1722–1733
Gbobaniyi E, Sarr A, Sylla MB, Diallo I, Lennard C, Dosio A, Dhiediou
Kamga A, Klutse NAB, Hewitson B, Nikulin G, Lamptey B (2013)
Climatology, annual cycle and interannual variability of precipita-
tion and temperature in CORDEX simulations over West Africa.
Int J Climatol 34:2241–2257. doi:10.1002/joc.3834
Giannini A, Saravanan R, Chang P (2003) Oceanic forcing of Sahel
rainfall on interannual to interdecadal timescales. Science
302:1027–1030
123
Environ Earth Sci (2017) 76:366
Giorgetta M, Wild M (1995) The water vapour continuum and its
representation in ECHAM4. MPI Rep. 162, p 38
Giorgi F, Jones C, Asrar GR (2009) Addressing climate information
needs at the regional level: the CORDEX framework. World
Meteorol Org Bull 58:175
Gregory D, Allen S (1991) The effect of convective downdraughts
upon NWP and climate simulations. In: 9th Conference on
Numerical Weather Prediction, Denver, CO, American Meteo-
rological Society pp 122–123
Gregory D, Rowntree PR (1990) A mass flux convection scheme with
representation of cloud ensemble characteristics and stability-
dependent closure. Mon Weather Rev 118:1483–1506
Grell GA (1993) Prognostic evaluation of assumptions used by
cumulus parameterizations. Mon Weather Rev 121:764–787
Hagemann S (2002) An improved land surface parameter dataset for
global and regional climate models. MPI Rep. 336, p 21
Harris I, Jones PD, Osborn TJ, Lister DH (2014) Updated high
resolution grids of monthly climatic observations—the CRU
TS3.10 dataset. Int J Clim 34:623–642. doi:10.1002/joc.3711
Hastenrath S (1990) The relationship of highly reflective clouds to
tropical climate anomalies. J. Climate 3:353–365
Herzog HJ, Vogel G, Schubert U (2002) LLM–a nonhydrostatic
model applied to high-resolving simulations of turbulent fluxes
over heterogeneous terrain. Theor Appl Climatol 73:67–86
Hijmans RJ, Cameron SE, Parra JL, Jones PG, Jarvis A (2005) Very
high resolution interpolated climate surfaces for global land
areas. Int J Climatol 25(15):1965–1978
Holtslag AAM, de Bruijn EIF, Pan HL (1990) A high resolution air
mass transformation model for short-range weather forecasting.
Mon Weather Rev 118:1561–1575
Huffman GJ, Adler RF et al (2001) Global precipitation at one-degree
daily resolution from multisatellite observations. J Hydrometeo-
rol 2(1):36–50
Jacob D (2001) A note to the simulation of the annual and interannual
variability of the water budget over the Baltic Sea drainage
basin. Meteorol Atmos Phys 77:61–73
Jacob D, Barring L, Christensen OB, Christensen JH, de Castro M,
Deque M, Giorgi F, Hagemann S, Hirschi M, Jones R,
Kjellstrom E, Lenderink G, Rockel B, Sanchez E, Schar C,
Seneviratne SI, Somot S, van Ulden A, van den Hurk B (2007)
An inter-comparison of regional climate models for Europe:
model performance in present day climate. Clim Change
81:31–52
Jenkins GS, Kamga A, Garba A, Diedhiou A, Morris V, Joseph E
(2002) Investigating the West African climate system using
global/regional climate models. Bull Am Meteorol Soc
83:583–595
Jones RG, Noguer M, Hassel D, Hudson D, Wilson S, Jenkins G,
Mitchell J (2004) Generating high resolution climate change
scenarios using PRECIS. Met Office Hadley Centre Handbook,
p 40. [Available online at http://www.metoffice.gov.uk/media/
pdf/6/5/PRECIS_Handbook.pdf.]
Jones C, Giorgi F, Asrar G (2011) The coordinated regional
downscaling experiment: CORDEX an international downscal-
ing link to CMIP5CLIVAR Exchanges 16: pp 34–39
Jung G, Kunstmann H (2007) High-resolution regional climate
modeling for the Volta region of West Africa. J Geophys Res
112:D23108. doi:10.1029/2006JD007951
Kain JS, Fritsch JM (1990) A one-dimensional entraining/detraining
plume model and its application in convective parameterization.
J Atmos Sci 47:2784–2802
Kain JS, Fritsch JM (1993) Convective parameterization for
mesoscale models: the Kain–Fritsch scheme. The Representation
of Cumulus Convection in Numerical Models, Meteor. Monogr.
No. 46, Am Meteorol Soc pp 165–170
Environ Earth Sci (2017) 76:366
Kalognomou EA, Lennard C, Shongwe M, Pinto I, Favre A, Kent M,
Hewitson B, Dosio A, Nikulin G, Panitz H-J, Büchner M (2013)
A diagnostic evaluation of precipitation in CORDEX models
over southern Africa. J Clim 26:9477–9506
Kiehl JT, Hack JJ, Bonan GB, Boville BA, Briegleb BP, Williamson
DL, Rasch PJ (1996) Description of the NCAR Community
Climate Model (CCM3). NCAR Tech. Note NCAR/TN-
4201STR, p 152
Kim J, Waliser DE, Mattmann C, Goodale C, Hart A, Zimdars P,
Crichton D, Jones C, Nikulin G, Hewitson B, Jack C, Lennard C,
Favre A (2014) Evaluation of the CORDEX-Africa multi-RCM
hindcast: systematic model errors. Clim Dyn 42:1189–1202
Kummerow C, Hong Y, Olson WS, Yang S, Adler RF, McCollum J,
Ferraro R, Petty G, Shin DB, Wilheit TT (2001) The evolution of
the Goddard profiling algorithm (GPROF) for rainfall estimation
from passive microwave sensors. J Appl Met 40:1801–1840
Kuo HL (1965) On formation and intensification of tropical cyclones
through latent heat release by cumulus convection. J Atmos Sci
22:40–63
Le Barb́e L, Lebel T, Tapsoba D (2002) Rainfall variability in West
Africa during the years 1950–90. J Clim 15(2):187–202
Li J, Barker HW (2005) A radiation algorithm with correlated-k
distribution. Part I: local thermal equilibrium. J Atmos Sci
62:286–309
Lohmann U, Roeckner E (1996) Design and performance of a new
cloud microphysics scheme developed for the ECHAM general
circulation model. Clim Dyn 12:557–572
Louis JF (1979) A parametric model of vertical eddy fluxes in the
atmosphere. Bound Layer Meteorol 17:187–202
Lu J, Delworth TL (2005) Ocean forcing of the late 20th century
Sahel drought. Geophys Res Lett 32:L22706. doi:10.1029/
2005GL023316
McKee TB, Doesken NJ, Kleist J (1993) The relationship of drought
frequency and duration of time scales. In: 8th Conference on
Applied Climatology, American Meteorological Society, Jan
17–23, 1993, Anaheim CA, pp 179–186
Mlawer EJ, Taubman SJ, Brown PD, Iacono MJ, Clough SA (1997)
Radiative transfer for inhomogeneous atmospheres: RRTM, a
validated correlated-k model for the longwave. J. Geophys. Res,
102 (D14), 16 663–16 682
Morcrette JJ, Smith L, Fouquart Y (1986) Pressure and temperature
dependence of the absorption in long wave radiation parameter-
izations. Beitr Phys Atmos 59:455–469
Moron V, Ward MN (1998) ENSO teleconnections with climate variability
in the European and African sectors. Weather 53:287–295
Nicholson SE, Entekhabi D (1986) The quasi –periodic behavior of
rainfall variability in Africa and its relationship to the southern
oscillation. Arch Meteorol Geophys Bio Klimatol series A
34:311–348
Nicholson SE, Kim J (1997) The relationship of the El Niño-southern
oscillation to African rainfall. Int J Climatol 17:117–135
Nikulin G, Jones C, Giorgi F, Asrar G, Buchner M, Cerezo-Mota R,
Christensen OB, Deque M, Fernandez J, Hansler A, van
Meijgaard E, Samuelsson P, Sylla MB, Sushama L (2012)
Precipitation climatology in an ensemble of CORDEX-Africa
regional climate simulations. J Climate 25(18):6057–6078
Omotosho JB, Abiodun BJ (2007) A numerical study of moisture build-
up and rainfall over West Africa. Meteorol Appl 14:209–225
Paeth H, Hall NM, Gaertner MA, Alonso MD, Moumouni S, Polcher
J, Ruti PM, Fink AH, Gosset M, Lebel T, Gaye AT, Rowell DP,
Moufouma-Okia W, Jacob D, Rockel B, Giorgi F, Rum-
mukainen M (2011) Progress in regional downscaling of west
African precipitation. Atmos Sci Lett 12(1):75–82. doi:10.1002/
asl306
Pal JS, Small EE, Eltahir EAB (2000) Simulation of regional-scale
water and energy budgets: representation of subgrid cloud and
Page 19 of 20 366
precipitation processes within RegCM. J Geophys Res
105(D24):29 579–29 594
Pal JS et al (2007) Regional climate modeling for the developing
world: the ICTP RegCM3 and RegCNET. Bull Am Meteorol
Soc 88:1395–1409
Parry ML, Canziani OF, Palutikof JP, van der Linden PJ, Hanson CE
(2007) Impacts Adaptation and Vulnerability Contribution of
Working Group II to the Fourth Assessment Report of the
Intergovernmental Panel on Climate Change Cambridge Univer-
sity Press Cambridge UK 2007
Patricola CM, Cook KH (2007) Dynamics of the West African
monsoon under mid-Holocene processional forcing: regional
climate model simulations. J Clim 20:694–716
Pu B, Cook KH (2010) Dynamics of the West African westerly jet.
J Clim 23:6263–6276
Pu B, Cook KH (2011) Role of the West African Westerly jet in Sahel
rainfall variations. J Clim. doi:10.1175/JCLI-D-11-00394.1
Rasch PJ, Kristjansson JE (1998) A comparison of the CCM3 model
climate using diagnosed and predicted condensate parameteri-
zations. J Clim 11:1587–1614
Rechid D, Raddatz TJ, Jacob D (2009) Parameterization of snow-free
land surface albedo as a function of vegetation phenology based
on MODIS data and applied in climate modelling. Theor Appl
Climatol 95:245–255
Redelsperger JL, Diongue A, Diedhiou A, Ceron JP, Diop M,
Gueremy JF, Lafore JP (2002) Multiscale description of a
Sahelian synoptic weather system representative of the West
African monsoon. Q J R Meteorol Soc 128:1229–1257
Redelsperger JL, Thorncroft CD, Diedhiou A, Lebel T, Parker DJ,
Polcher J (2006) African monsoon multidisciplinary analysis: an
international research project and field campaign. Bull Am
Meteorol Soc 87(12):1739–1746. doi:10.1175/BAMS-87-12-1739
Ritter B, Geleyn JF (1992) A comprehensive radiation scheme of
numerical weather prediction with potential application to
climate simulations. Mon Weather Rev 120:303–325
Rockel B, Will A, Hense A (2008) The regional climate model
COSMO-CLM (CCLM). Meteorol Z 17:347–348
Ropelewski CF, Halpert MS (1987) Global and regional scale
precipitation patterns associated with the El Niño/Southern
Oscillation. Mon Weather Rev 115:1606–1626
Ropelewski CF, Halpert MS (1989) Precipitation pattern associated
with the high index phase of the southern oscillation. J Clim
2:268–284
Rummukainen M (2010) State-of-the-art with regional climate
models. Wiley Interdiscip Rev Clim Change 1:82–96. doi:10.
1002/wcc8
Samuelsson P, Gollvik S, Ullerstig A (2006) The land-surface
scheme of the Rossby Centre regional atmospheric climate
model (RCA3).SMHI Rep. Met. 122, p 25
Samuelsson P, Jones CG, Willen U, Ullerstig A, Gollvik S, Hansson
U, Jansson C, Kjellstrom E, Nikulin G, Wyser K (2011) The
Rossby Centre Regional Climate model RCA3: model descrip-
tion and performance. Tellus 63A:4–23
Sass BH, Rontu L, Savijarvi H, Raisanen P (1994) HIRLAM-2
radiation scheme: Documentation and tests. SMHI HIRLAM
Tech. Rep. 16, p 43
Savijarvi H (1990) A fast radiation scheme for mesoscale model and
short-range forecast models. J Appl Meteorol 29:437–447
Schulz JP, Dumenil L, Polcher J, Schlosser CA, Xue Y (1998) Land
surface energy and moisture fluxes: comparing three models.
J Appl Meteorol 37:288–307
Smith RNB (1990) A scheme for predicting layer clouds and their
water content in a general circulation model. Q J R Meteorol Soc
116:435–460
Sultan B, Janicot S (2000) Abrupt shift of the ITCZ over West Africa
and intraseasonal variability. Geophys Res Lett 27:3353–3356
123
366 Page 20 of 20
Sultan B, Janicot S (2003) The West African monsoon dynamics Part
II: The ‘‘pre-onset’’ and ‘‘onset’’ of the summer monsoon. J Clim
16:3407–3427
Sundqvist H, Berge E, Kristjansson JE (1989) Condensation and
cloud parameterization studies with a mesoscale numerical
weather prediction model. Mon Weather Rev 117:1641–1657
Sylla MB, Diallo I, Pal JS (2013) West African monsoon in state-of
the-art regional climate models In Climate Variability—Re-
gional and Thematic Patterns Tarhule A (ed) ISBN: 980-953-
307-816-3
Taylor KE (2001) Summarizing multiple aspects of model perfor-
mance in a single diagram. J Geophys Res 106(D7):7183–7192
Thorncroft CD, Nguyen H, Zhang C, Peyrille P (2011) Annual cycle
of the West African monsoon: regional circulations and associ-
ated water vapour transport. Q J R Meteorol Soc
137(654):129–147
Tiedtke M (1989) A comprehensive mass flux scheme for cumulus
parameterization in large-scale models. Mon Weather Rev
117:1779–1800
123
Environ Earth Sci (2017) 76:366
Tompkins AM (2002) A prognostic parameterization for the sub grid
scale variability of water vapor and clouds in large-scale models
and its use to diagnose cloud cover. J Atmos Sci 59:1917–1942
Verseghy DL (2000) The Canadian Land Surface Scheme (CLASS):
its history and future. Atmos Ocean 38:1–13
Vizy EK, Cook KH (2001) Mechanisms by which Gulf of Guinea and
eastern North Atlantic sea surface temperature anomalies
influence African rainfall. J Clim 14:795–821
Vizy EK, Cook KH (2002) Development and application of a
mesoscale climate model for the tropics: influence of sea surface
temperature anomalies on the West African monsoon. J Geophys
Res 107(D3):4023. doi:10.1029/2001JD000686
Zadra A, Caya D, Cote J, Dugas B, Jones C, Laprise R, Winger K,
Caron LP (2008) The next Canadian Regional Climate Model.
Phys Can 64:75–83